Ausum Studio wrote:
Why the need to dispute? !
People suggesting DTML still has a place in Zope. You shouldn't need to learn two templating languages and a scripting language just to use one web framework ;-)
Page templates aren't meant to support logic, right?
No, and neither is DTML. But DTML makes it easier for you to think you can put application logic into it without being horribly burned later.
So, there you have something that page templates couldn't do: when you do need logic and content in a single method.
You should _never_ _ever_ have content and logic in a single method!
Furthermore, I'd appreciate that you help me on this: How would you perform the following using ZPT or PythonScript?
<dtml-in objectItems> <dtml-var sequence-item> </dtml-in>
Jeez, this is documented and painfully simple: <tal:i repeat="item here/objectItems"> <tal:x replace="item"/> </tal:i> Of course, you probably want to display that, so something like the following is more likely: <table> <tr tal:repeat="item here/objectItems"> <td tal:content="item">Sample content</td> </tr> </table> In a python script, it'd be: for item in context.objectItems(): print item() # this line might need ot be edited # depending on what the items are return printed Now, any other excuses why you think DTML should still exist? ;-) cheers, Chris
----- Original Message ----- From: "Chris Withers" <chrisw@nipltd.com> To: "Ausum Studio" <ausum_studio@hotmail.com> Cc: <zope@zope.org> Sent: Friday, February 14, 2003 5:56 AM Subject: Re: Java re-invents DTML :-)
Ausum Studio wrote:
Why the need to dispute? !
People suggesting DTML still has a place in Zope. You shouldn't need to learn two templating languages and a scripting language just to use one web framework ;-)
I must agree to that. Let's all improve DTML and trash ZPT ! ;) Talking seriously, after all this time of watching things going by, I've come to think that by having both languages we have an advantage over the competition. We'd need to reinforce that instead of *bashing* one of them.
Page templates aren't meant to support logic, right?
No, and neither is DTML. But DTML makes it easier for you to think you can put application logic into it without being horribly burned later.
So, there you have something that page templates couldn't do: when you do need logic and content in a single method.
You should _never_ _ever_ have content and logic in a single method!
In my opinion it depends on the size of the method and its purpose, as what does matter is how easy is to manage, read and reuse the code. At some place there might be the equation evaluating the code management efficiency (for a determined task), as a result of comparing the number of methods, its size in characters, and the changeability of the content that could possibly be inside them. I have my own equation, and find that yes, there are cases when a single method for a few tasks drives to an overall better code management.
Furthermore, I'd appreciate that you help me on this: How would you
perform
the following using ZPT or PythonScript?
<dtml-in objectItems> <dtml-var sequence-item> </dtml-in>
Jeez, this is documented and painfully simple:
I knew you'd say that! :)
<tal:i repeat="item here/objectItems"> <tal:x replace="item"/> </tal:i>
Of course, you probably want to display that, so something like the
following is
more likely:
<table> <tr tal:repeat="item here/objectItems"> <td tal:content="item">Sample content</td> </tr> </table>
In a python script, it'd be:
for item in context.objectItems(): print item() # this line might need ot be edited # depending on what the items are return printed
Now, any other excuses why you think DTML should still exist? ;-)
Unfortunately neither the three of them work :) (in the context of my needs, of course) When you have the time, I'd appreciate that you take a look at CMFPortlet's code, at the part I use to call portlet views. Lucky me that DTML did the job. ;) Ausum
Come on folks. There's nothing wrong with supporting two templating languages. Quite honestly, dtml is much easier to learn for most and has it's place in application development -- just as ZPT does. One simple reason that dtml should be supported (in addition to the fact that it works well) is for the support of existing Zope applications. If Zope was to suddenly stop supporting dtml, it would cause many, many people a lot of work if they wanted to get features available in a new version of Zope. The argument could be made that they "need" to migrate to zpt in the opinion of some, but that is not good business for Zope. I don't know why there are a zealous few out there that want to eliminate dtml from Zope. If you don't like it, don't use it. This is technology folks, not religion. Oh yeah, my bad -- for some, technology is religion. ;-) With the proper discipline, dtml or zpt along with python can help achieve good separation between logic and content. I've seen this same argument over the last 20 years whether the language was C, C++, Java, Prolog, Basic, Cobol, Pascal, etc. Trust me, crappy coding knows no language barrier. DTML and ZPT both have a place in Zope. Deciding which one to use should be left up to the individual. Kevin Chris Withers wrote:
Ausum Studio wrote:
Why the need to dispute? !
People suggesting DTML still has a place in Zope. You shouldn't need to learn two templating languages and a scripting language just to use one web framework ;-)
Page templates aren't meant to support logic, right?
No, and neither is DTML. But DTML makes it easier for you to think you can put application logic into it without being horribly burned later.
So, there you have something that page templates couldn't do: when you do need logic and content in a single method.
You should _never_ _ever_ have content and logic in a single method!
Furthermore, I'd appreciate that you help me on this: How would you perform the following using ZPT or PythonScript?
<dtml-in objectItems> <dtml-var sequence-item> </dtml-in>
Jeez, this is documented and painfully simple:
<tal:i repeat="item here/objectItems"> <tal:x replace="item"/> </tal:i>
Of course, you probably want to display that, so something like the following is more likely:
<table> <tr tal:repeat="item here/objectItems"> <td tal:content="item">Sample content</td> </tr> </table>
In a python script, it'd be:
for item in context.objectItems(): print item() # this line might need ot be edited # depending on what the items are return printed
Now, any other excuses why you think DTML should still exist? ;-)
cheers,
Chris
_______________________________________________ Zope maillist - Zope@zope.org http://mail.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope-dev )
Kevin Carlson wrote:
Come on folks. There's nothing wrong with supporting two templating languages. Quite honestly, dtml is much easier to learn for most and has it's place in application development -- just as ZPT does. ... DTML and ZPT both have a place in Zope. Deciding which one to use should be left up to the individual.
+1 I like DTML, because it's easy to use and to teach to beginners... and in many circumstances DTML is the only chance to do important things without hacking up some nasty "mixed-presentation/logic"-PythonScripts (e.g. generate formated reports and mails out of your ZODB, don't try this with ZPT...;)). If you've mastered the "namespace-thing" in DTML, it's a very flexible and powerful template-language. People should be encouraged to seperate design and logic (in any template-language), but not to drop DTML. The only argument for dropping DTML would be if ZC says it don't has the manpower to support it furthermore... If this would be the case, we should found a DTML-User-Group (the DUG) which will maintain DTML in the future...:) Cheers, Maik
Maik Jablonski wrote:
I like DTML, because it's easy to use and to teach to beginners...
Really? How about explaining all the weird name lookups and orders and (_.None,_)'s that you end up with?
in many circumstances DTML is the only chance to do important things without hacking up some nasty "mixed-presentation/logic"-PythonScripts
I realyl don't agree ;-)
(e.g. generate formated reports and mails out of your ZODB, don't try this with ZPT...;)).
Gimme an example problem and I'll suggest a solution... cheers, Chris
Kevin Carlson wrote:
Come on folks. There's nothing wrong with supporting two templating languages. Quite honestly, dtml is much easier to learn for most and has it's place in application development -- just as ZPT does.
Quite true. DTML can be very handy for templating non-tag-based text. On the other hand, DTML has several misfeatures and drawbacks that ought to be fixed, and leveraging TALES may be a good way to do it. For example, instead of: <dtml-with x><dtml-with y><dtml-var z></dtml-with></dtml-with> ...or... <dtml-var expr="x['y'].z()"> ...we could have... <dtml-var name="x/y/z"> Another advantage would be the ability to control context, writing "here/foo" to acquire a foo object and "request/foo" to get request variable "foo". This was proposed way back when, but nobody has had the time to do it :-( Cheers, Evan @ 4-am
On Fri, Feb 14, 2003 at 10:49:50AM -0600, Evan Simpson wrote: | Kevin Carlson wrote: | >Come on folks. There's nothing wrong with supporting two templating | >languages. Quite honestly, dtml is much easier to learn for most and | >has it's place in application development -- just as ZPT does. | | Quite true. DTML can be very handy for templating non-tag-based text. | | On the other hand, DTML has several misfeatures and drawbacks that ought | to be fixed, and leveraging TALES may be a good way to do it. For | example, instead of: | | <dtml-with x><dtml-with y><dtml-var z></dtml-with></dtml-with> | ...or... | <dtml-var expr="x['y'].z()"> | | ...we could have... | | <dtml-var name="x/y/z"> | | Another advantage would be the ability to control context, writing | "here/foo" to acquire a foo object and "request/foo" to get request | variable "foo". | | This was proposed way back when, but nobody has had the time to do it :-( Hey! I thought this was implemented/was planned to be implemented in Zope3. Anyone has an idea? []'s -- Sidnei da Silva (dreamcatcher) <sidnei@x3ng.com.br> X3ng Web Technology <http://www.x3ng.com.br> GNU/Linux user 257852 Debian GNU/Linux 3.0 (Sid) 2.4.18 ppc If it has syntax, it isn't user friendly.
-----Original Message----- From: zope-admin@zope.org [mailto:zope-admin@zope.org]On Behalf Of Evan Simpson Sent: Friday, February 14, 2003 8:50 AM To: zope@zope.org Subject: [Zope] Re: Java re-invents DTML :-)
Kevin Carlson wrote:
Come on folks. There's nothing wrong with supporting two templating languages. Quite honestly, dtml is much easier to learn for most and has it's place in application development -- just as ZPT does.
Quite true. DTML can be very handy for templating non-tag-based text.
On the other hand, DTML has several misfeatures and drawbacks that ought to be fixed, and leveraging TALES may be a good way to do it. For example, instead of:
<dtml-with x><dtml-with y><dtml-var z></dtml-with></dtml-with> ...or... <dtml-var expr="x['y'].z()">
...we could have...
<dtml-var name="x/y/z">
Another advantage would be the ability to control context, writing "here/foo" to acquire a foo object and "request/foo" to get request variable "foo".
This was proposed way back when, but nobody has had the time to do it :-(
Too bad. This would pretty nice. Would this be done as a third kind of DTML object? How would you prevent breaking older DTML objects?
On Friday 14 February 2003 12:52, Charlie Reiman wrote:
Too bad. This would pretty nice. Would this be done as a third kind of DTML object? How would you prevent breaking older DTML objects?
My DTML2 proposal for Zope 3 covers all of the issues you guys mentioned: http://dev.zope.org/Wikis/DevSite/Projects/ComponentArchitecture/DocumentTem... Regards, Stephan -- Stephan Richter CBU Physics & Chemistry (B.S.) / Tufts Physics (Ph.D. student) Web2k - Web Software Design, Development and Training
On Fri, 2003-02-14 at 11:02, Stephan Richter wrote:
My DTML2 proposal for Zope 3 covers all of the issues you guys mentioned: http://dev.zope.org/Wikis/DevSite/Projects/ComponentArchitecture/DocumentTem...
Regards, Stephan
Stephan, I like your proposal a lot. I chose Zope originally BECAUSE of dtml. I was not a hard core programmer and the idea of a tag based scripting language appealed to me since I was familiar with html. One question I do have, it seemed like you were proposing that there would be something to handle the old dtml1 in Zope 3, and there would also be the new implementation of dtml2 along with zpt etc. Am I understanding this correctly? And if so, could you run a site on Zope3 using dtml1 and then migrate it to dtml2 and would the code coincide peacefully together during that transistion process? Or, am I way off-base here? -- Michael Lewis <michael@nichestaffing.com> NicheStaffing.com
On Friday 14 February 2003 13:48, Michael Lewis wrote:
I like your proposal a lot.
Thanks.
I chose Zope originally BECAUSE of dtml. I was not a hard core programmer and the idea of a tag based scripting language appealed to me since I was familiar with html.
Right, I personally like DTML more too.
One question I do have, it seemed like you were proposing that there would be something to handle the old dtml1 in Zope 3, and there would also be the new implementation of dtml2 along with zpt etc. Am I understanding this correctly?
Yes, the old implementation of DTML will still be available.
And if so, could you run a site on Zope3 using dtml1 and then migrate it to dtml2 and would the code coincide peacefully together during that transistion process? Or, am I way off-base here?
Yes, since I think they will be two different object types or you will be able to explicitly select between the new and the old version. However, we have not thought at all about migration tools, but I imagine there will be some for the common cases, so that you can just run a script that attempts migrating DTML1 to DTML2 code. Regards, Stephan -- Stephan Richter CBU Physics & Chemistry (B.S.) / Tufts Physics (Ph.D. student) Web2k - Web Software Design, Development and Training
Stephan Richter wrote:
My DTML2 proposal for Zope 3 covers all of the issues you guys mentioned: http://dev.zope.org/Wikis/DevSite/Projects/ComponentArchitecture/DocumentTem...
Very nice! It *is* possible to compatibly retrofit TALES into much of DTML 1, though. See my languishing proposal at: http://dev.zope.org//Wikis/DevSite/Proposals/DTMLplusTALES Cheers, Evan @ 4-am
Evan Simpson wrote:
Quite true. DTML can be very handy for templating non-tag-based text.
I think ZPT does just fine with this and means you don't have to learn a second templating language.
<dtml-with x><dtml-with y><dtml-var z></dtml-with></dtml-with> ...or... <dtml-var expr="x['y'].z()">
...we could have...
<dtml-var name="x/y/z">
Another advantage would be the ability to control context, writing "here/foo" to acquire a foo object and "request/foo" to get request variable "foo".
All good points, but until someone can act on them, they don't really count ;-) cheers, Chris
At 06:23 AM 2/14/2003, Kevin Carlson wrote:
Come on folks. There's nothing wrong with supporting two templating languages. Quite honestly, dtml is much easier to learn for most and has it's place in application development -- just as ZPT does.
DTML and ZPT both have a place in Zope. Deciding which one to use should be left up to the individual.
Amen to that! I use both and find different things to like about each. The *real* villain here is the temptation to mix logic and presentation. Either approach will gracefully allow you to create unmaintainable code. Either approach will work well for those who show some discipline in their design. Let's spend more time encouraging good coding habits and less time blaming our tools, can we? Dylan
Kevin Carlson wrote:
Come on folks. There's nothing wrong with supporting two templating languages. Quite honestly, dtml is much easier to learn for most and has it's place in application development
...not unless some of its major current flaws are corrected. I think this is possible but I don't see the point of having two templating languages for one app server especially as we're now getting to the stage where you need to know _both_ before you can do anything useful with Zope :-(
One simple reason that dtml should be supported (in addition to the fact that it works well) is for the support of existing Zope applications. If Zope was to suddenly stop supporting dtml, it would cause many, many people a lot of work if they wanted to get features available in a new version of Zope. The argument could be made that they "need" to migrate to zpt in the opinion of some, but that is not good business for Zope.
Indeed. I think DTML should be supported for the rest of Zope 2.x. I really _don't_ think it should appear in Zope 3, but no-dobut someone has already shoved it in there looking for a quick fix for SQL or some such :-(
I don't know why there are a zealous few out there that want to eliminate dtml from Zope. If you don't like it, don't use it. This is technology folks, not religion. Oh yeah, my bad -- for some, technology is religion. ;-)
It's not religion, it's marketing. Zope currently has two templating languages. This splits the development effort for templating languages, means two sets of documentation and examples have to be kept up to date, two sets of bugs have to be looked after, etc. It also means new users have to face two sets of learning because no-one has the guts to clearly recommend one language over the other. That's not good for Zope. cheers, Chris
Hi, I'm a newbie who adopted Zope only because of ZPT. I've been following this discussion (as well as other readings) to see what DTML gives in addition to ZPT. Up to now, all I can see are the disadvantages you just pointed out. It has been quite an investment to learn zpt, python, acquisition, libraries, etc. Learning DTML just seems too much and so far I could not see the point of it, other than for reusing what exists (which is not a technical argument regarding DTML itself). Regards, Fernando Chris Withers wrote:
Kevin Carlson wrote:
Come on folks. There's nothing wrong with supporting two templating languages. Quite honestly, dtml is much easier to learn for most and has it's place in application development
...not unless some of its major current flaws are corrected. I think this is possible but I don't see the point of having two templating languages for one app server especially as we're now getting to the stage where you need to know _both_ before you can do anything useful with Zope :-(
One simple reason that dtml should be supported (in addition to the fact that it works well) is for the support of existing Zope applications. If Zope was to suddenly stop supporting dtml, it would cause many, many people a lot of work if they wanted to get features available in a new version of Zope. The argument could be made that they "need" to migrate to zpt in the opinion of some, but that is not good business for Zope.
Indeed. I think DTML should be supported for the rest of Zope 2.x. I really _don't_ think it should appear in Zope 3, but no-dobut someone has already shoved it in there looking for a quick fix for SQL or some such :-(
I don't know why there are a zealous few out there that want to eliminate dtml from Zope. If you don't like it, don't use it. This is technology folks, not religion. Oh yeah, my bad -- for some, technology is religion. ;-)
It's not religion, it's marketing. Zope currently has two templating languages. This splits the development effort for templating languages, means two sets of documentation and examples have to be kept up to date, two sets of bugs have to be looked after, etc. It also means new users have to face two sets of learning because no-one has the guts to clearly recommend one language over the other. That's not good for Zope.
cheers,
Chris
_______________________________________________ Zope maillist - Zope@zope.org http://mail.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope-dev )
On 02/19/03 00:09, Fernando Martins wrote:
Hi,
I'm a newbie who adopted Zope only because of ZPT. I've been following this discussion (as well as other readings) to see what DTML gives in addition to ZPT.
1. DTML can be used to generate output that isn't valid HTML/XML (e.g. if you have a standard "confirmation email" or standard email invoice you send out, you might write a DTML template for said email). 2. Practicality. There are some interesting products that are built with DTML. If you want to customize them, you might need to learn DTML.
Up to now, all I can see are the disadvantages you just pointed out. It has been quite an investment to learn zpt, python, acquisition, libraries, etc. Learning DTML just seems too much and so far I could not see the point of it, other than for reusing what exists (which is not a technical argument regarding DTML itself).
If you really needed it, you'd probably know. HTH -- Brad Bollenbach Software Engineer BBnet.ca
Brad Bollenbach wrote:
I'm a newbie who adopted Zope only because of ZPT. I've been following this discussion (as well as other readings) to see what DTML gives in addition to ZPT.
1. DTML can be used to generate output that isn't valid HTML/XML (e.g. if you have a standard "confirmation email" or standard email invoice you send out, you might write a DTML template for said email).
You can do this in ZPT, I've already written examples that are now in various mailing list archvies.
2. Practicality. There are some interesting products that are built with DTML. If you want to customize them, you might need to learn DTML.
This, sadly, is true :-S Even Squishdot is still scripted in DTML because I just can't stomach wading through it all to re-write in the ZPT ;-) cheers, Chris
Hi, I'm a newbie who adopted Zope only because of ZPT. I've been following this discussion (as well as other readings) to see what DTML gives in addition to ZPT. Up to now, all I can see are the disadvantages you just pointed out. It has been quite an investment to learn zpt, python, acquisition, libraries, etc. Learning DTML just seems too much and so far I could not see the point of it, other than for reusing what exists (which is not a technical argument regarding DTML itself). Regards, Fernando Chris Withers wrote:
Kevin Carlson wrote:
Come on folks. There's nothing wrong with supporting two templating languages. Quite honestly, dtml is much easier to learn for most and has it's place in application development
...not unless some of its major current flaws are corrected. I think this is possible but I don't see the point of having two templating languages for one app server especially as we're now getting to the stage where you need to know _both_ before you can do anything useful with Zope :-(
One simple reason that dtml should be supported (in addition to the fact that it works well) is for the support of existing Zope applications. If Zope was to suddenly stop supporting dtml, it would cause many, many people a lot of work if they wanted to get features available in a new version of Zope. The argument could be made that they "need" to migrate to zpt in the opinion of some, but that is not good business for Zope.
Indeed. I think DTML should be supported for the rest of Zope 2.x. I really _don't_ think it should appear in Zope 3, but no-dobut someone has already shoved it in there looking for a quick fix for SQL or some such :-(
I don't know why there are a zealous few out there that want to eliminate dtml from Zope. If you don't like it, don't use it. This is technology folks, not religion. Oh yeah, my bad -- for some, technology is religion. ;-)
It's not religion, it's marketing. Zope currently has two templating languages. This splits the development effort for templating languages, means two sets of documentation and examples have to be kept up to date, two sets of bugs have to be looked after, etc. It also means new users have to face two sets of learning because no-one has the guts to clearly recommend one language over the other. That's not good for Zope.
cheers,
Chris
_______________________________________________ Zope maillist - Zope@zope.org http://mail.zope.org/mailman/listinfo/zope ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope-dev )
On Friday 14 February 2003 12:56, Chris Withers wrote:
Now, any other excuses why you think DTML should still exist? ;-)
Errmm... When you use WYSIWYG like Macromedia DreamWeaver, it works with DTML without ANY problem and horribly brokes TAL tags that TAL pages becomes unusable at all. Clue: DTML still should exist at least for two points: - Graphish folks *likes* DreamWeaver because it is _best_ WYSIWYG yet - Macromedia would not like to GPL DreamWeaver and/or improve TAL support ;-) ...and my point: DTML is faster than TAL. :P -- Regards, Bogdan I used to think I was indecisive, but now I'm not so sure.
Bo M. Maryniuck wrote:
On Friday 14 February 2003 12:56, Chris Withers wrote:
Now, any other excuses why you think DTML should still exist? ;-)
Errmm... When you use WYSIWYG like Macromedia DreamWeaver, it works with DTML without ANY problem and horribly brokes TAL tags that TAL pages becomes unusable at all.
That's a point of view issue. Seeing the DTML tags in the preview in Dreamweaver counts as 'broken' for me, obviously not for you. What do you mean by 'horrible brokes TAL tags' though?
- Graphish folks *likes* DreamWeaver because it is _best_ WYSIWYG yet
...and it was one of the target audiences for ZPT, so if ZPT doesn't work with it, it's a bug that needs fixing.
...and my point: DTML is faster than TAL. :P
...then why aren't you writing your web server in raw assembler? cheers, Chris
On Monday 17 February 2003 18:57, Chris Withers wrote:
...then why aren't you writing your web server in raw assembler?
Ok-ok, I give up... Don't shoot, don't shoot... But I think ZPT is not a right tool for creating non-HTML stuff, like CSS, XML etc, do you? And ZPT is too explicit. OTOH, both are good and can coexist together. Are you agree here? ;-) -- Regards, Bogdan "On the Internet, no one knows you're using Windows NT" (Submitted by Ramiro Estrugo, restrugo@fateware.com)
Bo M. Maryniuck wrote:
But I think ZPT is not a right tool for creating non-HTML stuff, like CSS, XML etc, do you?
I think ZPT is fine for doing non-HTML stuff. It's specifically designed for geenrating XML, and it does pretty well with things like SQL and CSS too, especially when you consider it means you don't have to learn another templating language.
And ZPT is too explicit.
There is no such thing. If you don't like things explicit then you shouldn't be using something python-based. Explicitness also results in a lot less confusion in the long term...
OTOH, both are good and can coexist together. Are you agree here? ;-)
They can, I just want to avoid the situation where you _need_ to know both, as is currently the case. I'd also prefer to see new users advised to start with ZPT rather than having to struggle with DTML befoer they find ZPT. cheers, Chris
Chris Withers wrote:
Bo M. Maryniuck wrote: [...]
OTOH, both are good and can coexist together. Are you agree here? ;-)
They can, I just want to avoid the situation where you _need_ to know both, as is currently the case.
+1, that's exactly my main problem as I'm learning Zope.
I'd also prefer to see new users advised to start with ZPT rather than having to struggle with DTML befoer they find ZPT.
but don't worry about the newbies. The smart ones will start with ZPT :-) Cheers, Fernando
Fernando Martins wrote:
I'd also prefer to see new users advised to start with ZPT rather than having to struggle with DTML befoer they find ZPT.
but don't worry about the newbies. The smart ones will start with ZPT :-)
Yes, but they currently _will_ have to learn DTML to write ZSQL methods. That's a pretty big thorn in my side :-( Chris
From: Chris Withers [mailto:chrisw@nipltd.com] Yes, but they currently _will_ have to learn DTML to write ZSQL methods. That's a pretty big thorn in my side :-(
Hmm, indeed, it was a problem for me. When I start reading the zope book chapter on databases I was quickly derailed when I arrived to dtml stuff. I, naively (?), thought there was a way to do it without dtml... I'm living fine with <dtml-sqlvar> but maybe I'm missing a lot? Fernando
On Thursday 20 February 2003 18:55, Chris Withers wrote:
Yes, but they currently _will_ have to learn DTML to write ZSQL methods. That's a pretty big thorn in my side :-(
You suggest write SQL's with ZPT? 8-) -- Regards, Bogdan Bumper sticker: If you can read this, I can hit my brakes and sue you.
Bo M. Maryniuck wrote:
On Thursday 20 February 2003 18:55, Chris Withers wrote:
Yes, but they currently _will_ have to learn DTML to write ZSQL methods. That's a pretty big thorn in my side :-(
You suggest write SQL's with ZPT? 8-)
I have in the past adn I haven't had anyone persuade me otherwise... ...check the list archvies for examples. cheers, Chris
On Friday 21 February 2003 13:07, Chris Withers wrote:
I have in the past adn I haven't had anyone persuade me otherwise...
Hmm, I write logically complicated SQL's in PythonScript. This even better and faster (but not nice). -- Regards, Bogdan How much net work could a network work, if a network could net work?
Bo M. Maryniuck wrote:
On Friday 21 February 2003 13:07, Chris Withers wrote:
I have in the past adn I haven't had anyone persuade me otherwise...
Hmm, I write logically complicated SQL's in PythonScript. This even better and faster (but not nice).
That sounds like a good strategy. cheers, Chris
Chris Withers wrote:
...then why aren't you writing your web server in raw assembler?
You're backing up your one-sided argument with fallacious supposition. This seems to be a popular tendancy when issues of speed are discussed. It is not objective, it doesn't help the adoption of ZPT on its own merits, I would ask the community at large to cease this flawed line of reasoning. -- Jamie Heilman http://audible.transient.net/~jamie/ "...thats the metaphorical equivalent of flopping your wedding tackle into a lion's mouth and flicking his lovespuds with a wet towel, pure insanity..." -Rimmer
On Mon, Feb 17, 2003 at 12:42:45PM -0800, Jamie Heilman wrote:
Chris Withers wrote:
...then why aren't you writing your web server in raw assembler?
You're backing up your one-sided argument with fallacious supposition. This seems to be a popular tendancy when issues of speed are discussed. It is not objective, it doesn't help the adoption of ZPT on its own merits, I would ask the community at large to cease this flawed line of reasoning.
May I +1 ? Jerome Alet
participants (15)
-
Ausum Studio -
Bo M. Maryniuck -
Brad Bollenbach -
Charlie Reiman -
Chris Withers -
Dylan Reinhardt -
Evan Simpson -
Fernando Martins -
Jamie Heilman -
Jerome Alet -
Kevin Carlson -
Maik Jablonski -
Michael Lewis -
Sidnei da Silva -
Stephan Richter